The scientific revolutions of the last two decades in molecular biology and computer science have created the basis for greatly increasing our understanding of cancer etiology. In this program project we utilize approaches derived from these fields to study the etiology of three common cancers and their precursor lesions: 1) the adenomatous polyp and colon cancer, 2) dysplastic nevi and melanoma, and 3) proliferative breast disease and breast cancer. The underlying theme of the program project is that the analysis of precursor lesions to common cancers is a major key to understanding their etiology. Familial clusters of disease are a major focus of the project; they permit detailed analysis of genetic susceptibility, provide the basis for gene mapping, and provide a high risk population for analysis of other risk factors. The details of the research project for each trait vary according to our current level of understanding, but three common goals exist for all three studies consisting of: 1) definition and genetic analysis of precursor lesions of common cancers. 2) analysis of heritable and environmental factors which affect the expression of the precursor phenotype, and 3) analysis of factors which affect the conversion of the precursor lesion to malignancy. Cumulatively these studies amount to a continual refinement of the genotypic description of the susceptibility loci. We have made considerable progress defining the precursor lesions for colon cancer and melanoma, although these descriptions will be greatly furthered by project. The definition of genetic precursors for breast cancer is in the initial stage. By studying sites with a variable level of knowledge of the phenotype and role of the precursor, we can rapidly apply experience gained in one project to the others utilizing the infrastructure common to all the projects.